pyriemann.utils.covariance.normalize(X, norm)

Normalize a set of square matrices, using corr, trace or determinant.

Xndarray, shape (…, n, n)

The set of square matrices, at least 2D ndarray. Matrices must be invertible for determinant-normalization.

norm{“corr”, “trace”, “determinant”}

The type of normalization:

  • ‘corr’: normalized matrices are correlation matrices, with values in [-1, 1] and diagonal values equal to 1;

  • ‘trace’: trace of normalized matrices is 1;

  • ‘determinant’: determinant of normalized matrices is +/- 1.

Xnndarray, shape (…, n, n)

The set of normalized matrices, same dimensions as X.